Optimization of Bank Investment Portfolio and Debt Structure Under Sustainable Finance Policies Based on Mathematical Modeling Analysis

Guodong Zhang ^*

• Wuhan University, Wuhan, China

The increasing emphasis on sustainable finance policies has necessitated the development of advanced mathematical models to optimize bank investment portfolios and debt structures.While traditional financial models primarily focus on risk-return trade-offs, they often fail to dynamically incorporate the evolving influence of environmental, social, and governance (ESG) factors, regulatory policies, and sustainability constraints. Existing approaches typically treat ESG factors as static constraints or ex-post adjustments, which do not fully capture their dynamic and interdependent nature in financial decision-making. This study addresses these limitations by proposing a novel multi-objective optimization framework that integrates ESGadjusted risk-return dynamics, regulatory compliance constraints, and policy-driven investment incentives. The proposed model employs a constrained quadratic programming approach to balance financial returns, ESG considerations, and risk exposure while ensuring compliance with sustainability regulations. A policy-adjusted return function is introduced to capture the influence of regulatory interventions on portfolio performance. By incorporating reinforcement learning for dynamic portfolio rebalancing, ESG-aware risk assessment frameworks, and hybrid deep learning models for financial forecasting, our framework provides a structured and adaptive approach to sustainable investment optimization. Experimental simulations demonstrate the model's effectiveness in enhancing financial resilience, mitigating greenwashing risks, and optimizing debt structures under evolving regulatory environments. These findings offer valuable insights for policymakers and financial institutions, contributing to a more stable and sustainable financial system.